Personally I think that copying golf holes exactly is a bad idea unless you are doing it for some particularly reason. An example of that might be what some would like to do on Gulph Mills's #10 by moving the present green about 60yds to the right to make the hole play better and more reasonable. The only problem with that is the back 2/3 of the present green is one of the best Perry Maxwell's anyone has ever seen! Although moving the green may make the hole a bit better doing so is just not worth sacrificing an original Maxwell green (even if it can be exactly duplicated) and I'm very much opposed to this alteration!
But exactly duplicating a hole or holes is not good either, in my opinion. I don't really see the enduring interest in playing NGLA's redan if it was duplicated in California.
MacDonald and Raynor had an interesting modus operandi of duplicating a number of their holes, particularly par 3s and I don't think that's a trend that should be aspired to. Actually the interest to me with MacDonald and Raynor's dupication is not in how similar their holes are but in their subtle and sometimes not so subtle differences and what that means to the golf on them!
I do though very much see the benefit and interest in doing what's called "conceptual copies" of various holes. That to me is when many of the shot concepts, shot demands and maybe the shot values are similar but that the holes are unrecognizable as similar. That may include things like using similar slope for the same shot requirement and maybe using a quarry in place of a creek for the same requirement or option. The shot demands may be very similar but similarity in look is unrecognizable.
I also don't really think that it's necessary to try to come up with something that is wholy new in golf architecture just for the sake of coming up with something new! There's enough in golf architecture (features) that's usable to come up with a multipicity of combinations that can go far enough to be considered unique anyway. In this way it probably isn't much different from chess that once past the first few moves the possibilities are endless (although sometimes questionable).
I see nothing wrong with MacDonaldization either if you happen upon natural landforms that are very similar. For instance, to date, I've seen about 4-5 natural landforms (raw land not golf holes) that are incredibly similar to Merion's #5. Some may have the creeks or the slope on the other side but still they are basically the same or certainly would create the same basic demands and shot values for the golfer. And if these natural landforms are ever used for golf I think they should be used as is, just like Merion's #5 was!
The golfer himself and what he can and can't do with a golf ball is probably far more limiting to golf architecture than architecture itself anyway!
I'm sort of interested too in the basic starting points of various architects and how they go about creating or producing things. Some clearly look very carefully at the site and different areas of the land and are able to imagine the natural possibilities on it to get the golfer to think and do various interesting things.
Other architects may not even notice the natural possibilities or certainly nowhere near as much as others and just decide that they need to make the golfer do this or that here or there and if they can't or don't find it they just make it, sometimes from scratch.
I think that certain formulaics of architecture do that to them too, like a 7000yd par 70 or any other various formulaic need.
But anyway attempting to copy is a bad idea, particularly if it looks similar.